超级电容器储能系统中DC/DC变换器先进PID控制改进方法  

An Improved Method of Advanced PID Control for DC/DC Converter of Super-capacitor Energy Storage System

作　　者：赵爽[1] 张步涵[1] 陈奕[1] 代晓康[1] 

机构地区：[1]华中科技大学强电磁工程与新技术国家重点实验室,湖北武汉430074

出　　处：《水电能源科学》2014年第6期176-179,共4页Water Resources and Power

基　　金：国家重点基础研究发展计划(973计划)项目(2009CB219702;2010CB227206)

摘　　要：针对超级电容器储能系统中的DC/DC变换器为高阶、非线性的系统,采用传统的PID控制难以应对负载、电压突变等复杂情况,提出了一种将Fletcher-Reeves共轭梯度法控制的BP神经网络控制器与PID相结合的先进PID控制改进方法,解决了DC/DC变换器传统控制算法中稳态误差大、控制响应时间长的问题。同时也建立了微网模型,并应用改进算法进行了仿真。仿真结果表明,所提出的改进方法能够有效地改善DC/DC变换器端电压的控制效果,使超级电容器储能系统能有效地平抑微网在并网状态下PCC点的功率波动。Owing to high-order nonlinear system for the DC/DC converter of super-capacitor energy storage system, the traditional PID control is difficult to cope with some complex situations, such as sudden change of load and voltage. Combined BP neural network controller of Fletcher-Reeves conjugate gradient method with PID control, an advanced PID control method is proposed to solve the problem of large steady-state error and long response time in the DC/DC converter with traditional control method. At the same time, the micro-network model is established and simulated with an improved algorithm. Simulation results show that an modified method can effectively improve the control effect of DC/DC converter＇s terminal voltage so that the super-capacitor energy storage system can effectively stabilize the micro-network＇ s power fluctuations at the PCC point.

关 键 词：超级电容器储能系统 BP神经网络 先进PID控制法 DC DC变换器 

分 类 号：TM743[电气工程—电力系统及自动化]